This invention relates to an improved thermowell or fluid sampling probe for use in chemical processing vessels, pipelines and the like.
Gas sampling probes, for example the insertion type, wherein a sample of gas has to be dynamically taken from a pipeline or large vessel, are well known but suffer from a number of problems due to the flowing nature of the fluids to be sampled and the required length of the probe.
There are a number of problems associated with thermowell probes and gas sampling probes for use with natural gas pipelines. For example, in designing such probes to meet the mechanical requirements of the installation may result; in a probe that has a large volume and generates significant turbulence. This again is incompatible with sampling requirements. Thus, such probes typically suffer from the following drawbacks: they have a large internal volume, which is incompatible with ‘real time’ analysis and environmental considerations, they are prone to inaccurate sampling (due to turbulence), and mechanical failure of the probe can result due to resonance failure that are a consequence of vortex shedding. These three drawbacks are described more fully below.
Firstly, following recognized guidelines for sampling natural gas, such as ISO 10715:2001, which states that samples should be taken from the middle ⅓ of the pipe, results in a “long” sample probe. Not only does the probe have to be at least ⅓ the diameter of the pipe (pipe size is often 2 ft-4 ft in diameter/600 mm to 1200 mm) but also the length has to be sufficient to connect the probe via a branch tee and flange or if permitted, by a threadolet. (Normally branch flanges are the preferred connection type). In many cases the length of a gas sampling probe is significantly, or even hugely increased by the requirement, for a retractable and isolatable probe. In this case the probe is connected by a branch tee, valve and flange combination.
Secondly, there is the need to consider the phenomena of vortex shedding and the possibility that the vortex shedding frequency may coincide with the natural frequency of the probe. Should the two coincide then it is very likely that the probe will fail (snap off) due to resonance effects.
The combination of the two points above forces a probe design of a fattish nature. (Normally a probe with about a 25 mm (1″) outside diameter). Due to the way tubes and pipes are manufactured, it is not economical/normal, to manufacture a tube of say 1″ OD (25 mm) with an ID of less than ½″ (12.5 mm).
In the case of gas sampling probes the combination of the ‘long’ length of the probe combined with the ‘relatively’ large internal diameter results in a significant gas hold up volume in the sample probe itself. This stored gas is often known as ‘dead space’ gas and has to be vented, or otherwise disposed of, before actual gas from the pipeline can enter the analyzer. The volume of stored or ‘dead space’ gas within the probe is further increased by the effect of pressure. For each bar of the pressure that the pipeline operates above atmospheric pressure, then the real (or normal or standard) volume of gas in the probe is increased by that ratio. For example if the internal volume of the probe was say 0.25 liters and the pressure of the pipeline it is operating in is 40 bara then the real (normal or standard) volume of stored or ‘dead gas’ within the probe will be approximately 0.25×40=10 liters. It is not uncommon for gas pipelines to be operating at 80 bara or even higher.
Thus, there is a problem designing a gas sampling probe with a response time fast enough to match an associated analytical system. In such circumstances a significant amount, of gas that has to be moved out of the way (vented) before a representative sample of the actual gas in the pipeline can be presented/introduced to the analyzer/sample cylinder connected to the sample probe. This venting process can be very damaging to the environment.
An alternative to using a pipe or tube would be to use a solid bar with a small hole ‘drilled’ down the middle. However, drilling a 2, 3, 4, 5 mm diameter hole or even larger, down the length of a stainless steel bar of typically say 0.3 to 2.0 meters long is no easy or cheap task. Additionally, the quality of the surface finish of such a drilled hole is difficult to control which brings its own problems to representative sampling of natural gas, especially with the higher hydrocarbons and reactive components.
Lastly, by introducing such a large protrusion into the flowing gas creates significant, turbulence which in turn can momentarily alter the composition of the gas. Small droplets of hydrocarbon liquid may be formed, similar to the white vapor trails often seen behind an airplane (except in the case of the airplane it is waiter droplets not: hydrocarbon liquid droplets). These small droplets not only change the gaseous phase composition but also have the potential to absorb, momentarily, any reactive components such as hydrogen sulphide. Therefore at the point in space (actually the point in the pipeline at the tip or entrance to the sample probe) where the gas is sampled from, every effort needs to be made to reduce the turbulence.
An object of this invention is to reduce the internal volume of a gas sampling probe. Another object of the invention is to minimize or eliminate vortex shedding induced by use of such a probe. A further object of the invention is to minimize the turbulence at the sampling point.